JP4734952B2 - Image processing method and apparatus - Google Patents

Description

  The present invention relates to an image processing apparatus used for a game machine or the like, and more particularly, to an image processing method and apparatus for improving display of an edge portion where a color change in a character image is severe.

  As is well known, image display methods using sprites are widely used in pachinko and other game machines. FIG. 4 is a diagram showing the configuration of a display system using the sprite display method. In this figure, 1 is a CPU (central processing unit), 2 is an image processing device, and 3 is a pattern in which sprite patterns are stored in advance. ROMs 4 and 4 are display devices such as a liquid crystal display. In such a configuration, the image processing apparatus 2 receives sprite attribute data indicating the number of sprites to be displayed, the display position, rendering process parameters, and the like from the CPU 1 and writes them in the internal sprite attribute table. Here, the rendering process parameter is a parameter for instructing a rendering process such as enlargement, reduction, or deformation of a sprite. The image processing device 2 reads the sprite pattern from the pattern ROM 3 based on the attribute data written in this table, performs processing such as enlargement / reduction based on the rendering processing parameters, and then outputs to the display device 4 for display. .

  By the way, in the above-described image display using sprites, if the sprite pattern is enlarged, the number of display dots increases from the original image. Therefore, it is necessary to determine the color of the increased display dots based on the color of the dots of the original image. Conventionally, a bilinear filter interpolation process is known as a process for determining the color of the increased display dots.

FIG. 5 is a diagram for explaining bilinear filter interpolation processing accompanying enlargement processing. In this figure, rgb (x, y) indicates the color of display dots (coordinates (x, y)) increased after enlargement. Color data to be used. Also, rgb (m, n), rgb (m + 1, n), rgb (m, n + 1), and rgb (m + 1, n + 1) are calculated display positions when the adjacent four dots before enlargement are enlarged (m , N), (m + 1, n), (m, n + 1), and (m + 1, n + 1) color data. These display positions are the calculation dots closest to the coordinates surrounding the coordinates (x, y) of the display dots. Note that m and n are integers. In the figure, p and q are values calculated from the coordinates (x, y) and the coordinates (m, n), (m + 1, n), (m, n + 1), (m + 1, n + 1) by the following equation. .
p = coordinate value x-coordinate value m
q = coordinate value y-coordinate value n

And rgb (x, y) is calculated | required by following Formula.
rgb (x, n) = (1−p) × rgb (m, n) + p × rgb (m + 1, n)
rgb (x, n + 1) = (1-p) × rgb (m, n + 1) + p × rgb (m + 1, n + 1)
rgb (x, y) = (1-q) * rgb (x, n) + q * rgb (x, n + 1)

In addition, the applicant of the present application as an image display device using bilinear filter interpolation processing,
Japanese Patent Application No. 2004-193019
Has been filed.

By the way, the above-described bilinear filter interpolation processing has the following problem at the edge portion where the color change of the character is severe. In other words, if you enlarge an image with relatively clear edges of adjacent images, such as an animation or a geometric picture, and set a new dot color by bilinear filter interpolation processing, the resulting image becomes the original image. On the other hand, the boundary portion of the character is particularly blurred. This tendency is particularly remarkable in the case of images such as animation.
Therefore, an object of the present invention is to provide an image processing method and apparatus capable of maintaining the sharpness of the edge of an image even after the image enlargement process.

The present invention has been made in order to solve the above-described problems. The present invention includes storage means for storing threshold data for obtaining color data of display dots to be displayed in advance for each sprite pattern. An image processing method in an image processing apparatus for displaying after performing a predetermined rendering process, wherein the display coordinates of display points to be displayed are calculated and the display coordinates of display points to be actually displayed are obtained by the rendering process of the sprite pattern. A first calculation process to be calculated; a second calculation process for calculating a color data difference between four display points in calculation around the display dots to be actually displayed; and the difference is stored in the storage means. A comparison process for detecting whether the stored sprite pattern to be displayed is smaller than the threshold data; and the comparison process Based on the output, when the difference in color data between adjacent display points out of the four display points is larger than the threshold data, the display dots to be actually displayed among the adjacent display points. Sets and outputs a calculation coefficient that uses color data at close display points as interpolation data, and if small, sets a calculation coefficient that uses color data obtained by bilinear interpolation of color data between adjacent display points as interpolation data And an interpolation process for obtaining color data of the display dots to be actually displayed using the calculation coefficient output from the calculation coefficient determination process .

The present invention includes storage means for storing threshold data for obtaining color data of display dots to be displayed in advance for each sprite pattern, and performs image processing for display after performing a predetermined rendering process on the sprite pattern. An image processing method in the apparatus, wherein a first display processing for obtaining display coordinates of a display dot to be actually displayed and a display coordinate of a display dot to be actually displayed are obtained by the rendering processing of the sprite pattern, and the actual display A second calculation process for calculating a difference in color data between four display points in calculation around a display dot to be calculated, and a difference between each difference in color data between the four display points and threshold data There is a pattern determined in advance by the relationship, and the difference obtained by the second calculation process is stored in the storage means to be displayed. Based on whether or not the threshold value data of the light pattern is smaller, a comparison process for detecting which one of the patterns corresponds, a pattern detected by the comparison process, and each preset pattern From the calculation coefficient correspondence process, the calculation coefficient determination process for calculating the calculation coefficient corresponding to the interpolation method to be applied among the nearest neighbor interpolation, the bilinear interpolation, and the bilinear-nearest neighbor mixed interpolation, and the calculation coefficient determination process And interpolation processing for obtaining color data of the display dots to be actually displayed using the output calculation coefficient .

The present invention is an image processing apparatus that displays after performing a predetermined rendering process on a sprite pattern, and threshold data for obtaining color data of display dots to be displayed is stored in advance for each sprite pattern. Storage means; first calculation means for obtaining display coordinates of display dots to be actually displayed along with calculation of display coordinates of display points by rendering processing of the sprite pattern; and display dots to be actually displayed Second calculation means for calculating the color data difference between the four display points in the surrounding calculation, and the threshold value data of the sprite pattern to be displayed stored in the storage means. Color data between adjacent display points among the four display points based on the output of the comparison means for detecting whether or not it is small When the difference is larger than the threshold value data, among the adjacent display points, an arithmetic coefficient is set and output with the color data of the display point closer to the display dot to be actually displayed as interpolation data. In the case of the calculation coefficient decoder that sets and outputs a calculation coefficient that uses color data obtained by bilinear interpolation calculation of color data between adjacent display points as interpolation data, and the calculation coefficient output from the calculation coefficient decoder And an interpolation circuit for obtaining color data of display dots to be actually displayed .

The present invention is an image processing apparatus that displays after performing a predetermined rendering process on a sprite pattern, and threshold data for obtaining color data of display dots to be displayed is stored in advance for each sprite pattern. Storage means; first calculation means for obtaining display coordinates of display dots to be actually displayed along with calculation of display coordinates of display points by rendering processing of the sprite pattern; and display dots to be actually displayed Second calculation means for calculating the difference in color data between the four display points in the surrounding calculation, and the magnitude relationship between each of the color data differences between the four display points and the threshold data. And the difference obtained by the second calculation means is stored in the storage means and the sprite pattern to be displayed is displayed. Based on whether the value is smaller than the threshold value data, the comparison means for detecting which of the patterns corresponds to the pattern detected by the comparison means, and the correspondence between each pattern set in advance and the calculation coefficient From the table, among the nearest neighbor interpolation, the bilinear interpolation, and the bilinear-nearest neighbor mixed interpolation, the computation coefficient decoder for obtaining a computation coefficient corresponding to the interpolation method to be applied, and the computation coefficient output from the computation coefficient decoder And an interpolation circuit for obtaining color data of display dots to be actually displayed .

  According to the present invention, there is an effect that the sharpness of the edge of the image can be maintained even after the image enlargement processing.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. 1 is a block diagram showing a configuration of a bilinear filter unit BU in an image processing apparatus according to an embodiment of the present invention, FIG. 2 is a diagram for explaining processing contents of a Δ-comparator 31 in FIG. 1, and FIG. FIG. 2 is a block diagram illustrating an overall configuration of an image processing apparatus.

  In FIG. 3, reference numeral 3 denotes a pattern ROM, which stores a plurality of sprite patterns compressed. Reference numeral 13 denotes a sprite attribute table, in which sprite attribute data is registered by a CPU (not shown). Here, the sprite attribute data includes the storage address of the sprite pattern in the pattern ROM 3, rendering processing parameters for instructing the enlargement, reduction, rotation, deformation, etc. of the pattern, data for instructing the display position, Δ data (described later), etc. There is. A decode controller 14 controls the decoding (decompression) processing of the compressed sprite pattern in the pattern ROM 3. A ROM interface 15 outputs a read address to the pattern ROM 3 and outputs a sprite pattern read from the pattern ROM 3 to the decoder 16.

  The decoder 16 decodes (expands) the sprite pattern read from the pattern ROM 3 and outputs it to the sprite buffer interface 17. The sprite buffer interface 17 develops the sprite pattern output from the decoder 16 in the sprite buffer 18. In response to an instruction from the rendering engine 21, the sprite pattern in the sprite buffer 18 is read and output to the rendering engine 21. Here, the sprite buffer 18 has a capacity capable of developing a plurality of sprite patterns. A rendering controller 20 reads the sprite attribute data in the sprite attribute table 13 and outputs the read attribute data to the rendering engine 21.

  The rendering engine 21 performs the rendering process of the sprite pattern read from the sprite buffer 18 according to the instruction of the rendering controller 20, and then performs the bilinear filter interpolation process of the sprite pattern by the bilinear filter unit BU of FIG. The processed data is output to the frame buffer interface 22.

  The frame buffer interface 22 draws the frame buffer 23 based on the sprite pattern output from the rendering engine 21. Further, the frame buffer interface 22 receives an instruction from the display controller 25, reads display data from the frame buffer 23, and outputs the display data to the display controller 25. The frame buffer 23 is a memory in which pattern data is written in correspondence with the display dots of the display device 4, and has a double buffer configuration for drawing / display. The display controller 25 generates various timing signals for image display and outputs them to the display device 4, and outputs the display data read from the frame buffer 23 to the display device 4 in synchronization with the timing signals. .

Next, the bilinear filter unit BU shown in FIG. 1 will be described with reference to FIG.
In FIG. 5, rgb (x, y) is color data indicating the color of the display dots (coordinates (x, y)) increased after enlargement, and rgb (m, n), rgb (m + 1, n). , Rgb (m, n + 1), rgb (m + 1, n + 1) are calculated display positions (m, n), (m + 1, n), (m, n + 1) when 4 dots before enlargement are enlarged. , (M + 1, n + 1) color data.
In FIG. 1, reference numeral 31 denotes a Δ-comparator that performs the following processes. First, among the respective display positions (m, n), (m + 1, n), (m, n + 1), and (m + 1, n + 1) in FIG. 5, R (red) and G (green) of adjacent dots. , B (blue) are obtained by the following equations.
R (m + 1, n) -R (m, n) = ΔR1
G (m + 1, n) -G (m, n) = ΔG1
B (m + 1, n) -B (m, n) = ΔB1

R (m, n + 1) -R (m, n) = ΔR2
G (m, n + 1) -G (m, n) = ΔG2
B (m, n + 1) -B (m, n) = ΔB2

R (m + 1, n + 1) −R (m, n + 1) = ΔR3
G (m + 1, n + 1) -G (m, n + 1) = ΔG3
B (m + 1, n + 1) −B (m, n + 1) = ΔB3

R (m + 1, n + 1) -R (m + 1, n) = ΔR4
G (m + 1, n + 1) −G (m + 1, n) = ΔG4
B (m + 1, n + 1) −B (m + 1, n) = ΔB4

Next, the following conditional expressions (1) to (4) are obtained by checking whether or not each of the above-described difference data is smaller than the Δ data (threshold value) set in the sprite attribute table 13. Check whether or not holds. “&&” means “and”.
(| ΔR1 | <Δ) && (| ΔG1 | <Δ) && (| ΔB1 | <Δ) (1)
(| ΔR2 | <Δ) && (| ΔG2 | <Δ) && (| ΔB2 | <Δ) (2)
(| ΔR3 | <Δ) && (| ΔG3 | <Δ) && (| ΔB3 | <Δ) (3)
(| ΔR4 | <Δ) && (| ΔG4 | <Δ) && (| ΔB4 | <Δ) (4)

Next, it is detected which pattern of A to L shown in FIG. 2 corresponds to the target dot (coordinates (x, y)) based on whether or not the conditional expressions (1) to (4) are satisfied.
Here, the pattern A is a case where all of the conditional expressions (1) to (4) are not established, and each of the four dots around the target dot (coordinates (x, y)) is greatly different. is there. Pattern B is a case where conditional expressions (1) and (3) are satisfied and (2) and (4) are not satisfied. The colors of the left and right dots are not so different, but the colors of the upper and lower dots are This is a big difference. Similarly, in the pattern C, when the conditional expressions (1) and (3) are not satisfied and (2) and (4) are satisfied, the pattern D is not satisfied with the conditional expressions (1), (3), and (4), ( This is a case where 2) is established, and the same applies to other patterns. Pattern L is a case where conditional expressions (1) to (4) are all satisfied.

The Δ-comparator 31 converts the pattern data PD indicating the pattern detected by the above-described process into color data rgb (m, n), rgb (m + 1, n), rgb (m, n + 1), rgb (m + 1, n + 1). ) And the operation coefficient decoder 32.
The arithmetic coefficient decoder 32 and the coordinates (x, y) obtained by the rendering process and the coordinates (m, n), (m + 1, n), (m, n + 1), (m + 1, n + 1) of four points surrounding the point. ) To obtain data p and q shown in FIG. Next, new interpolation processing data p ′ and q ′ are determined based on the obtained data p and q and the pattern data PD.

That is, when the pattern data PD is data indicating the pattern A, the color of the dot (nearest neighbor) closest to the target dot (coordinates (x, y)) among the four points is the same color. P ′ and q ′ are defined. For example,
p> 0.5 q <0.5
When the closest point to the target dot (coordinates (x, y)) is a point of coordinates (m + 1, n) as shown in FIG.
p '= 1
q ′ = 0
It is determined.

When the pattern data PD is data indicating the pattern B, bilinear interpolation similar to the conventional one is performed in the horizontal direction, and nearest neighbor interpolation is performed in the vertical direction. That is,
p '= p
q ′ = 0 (when q ≦ 0.5) or q ′ = 1 (when q> 0.5)
It is determined.

When the pattern data PD is data indicating the pattern C, bilinear interpolation similar to the conventional one is performed in the vertical direction, and nearest neighbor interpolation is performed in the horizontal direction. That is,
p ′ = 0 (when p ≦ 0.5) or q ′ = 1 (when p> 0.5)
q ′ = q
It is determined.

The same applies to the other pattern data PD. When the pattern data PD is data indicating the pattern L, bilinear interpolation similar to the conventional one is performed in the vertical and horizontal directions. That is,
p '= p
q ′ = q
It is determined.
The arithmetic coefficient decoder 32 outputs the data p ′ and q ′ obtained by the above-described process to the bilinear filter interpolation circuit 33.
The above-described data p ′ and q ′ are actually obtained by creating a correspondence table between the patterns A to L and the data p ′ and q ′ in advance.

The bilinear filter interpolation circuit 33 receives the above-described data p ′ and q ′ and obtains the color data rgb (x, y) of the target dot by the following equation.
rgb (x, n) = (1−p ′) × rgb (m, n) + p ′ × rgb (m + 1, n)
rgb (x, n + 1) = (1-p ′) × rgb (m, n + 1) + p ′ × rgb (m + 1, n + 1)
rgb (x, y) = (1−q ′) × rgb (x, n) + q ′ × rgb (x, n + 1)

  As described above, according to the bilinear filter unit BU, when the difference between the color data of the upper, lower, left and right dots surrounding the target dot (coordinates (x, y)) is smaller than the threshold data Δ, it is the same as the conventional case. However, if it is large, it is assumed that there is an edge portion where the character color changes drastically, and the color of the target dot is the same as the color of the nearest neighbor dot. As a result, it is possible to prevent blurring of the edge portion where the color change of the character is severe.

  By the way, normally, R, G, and B data are 8 bits, and therefore the selection range of the data Δ is 1 to 256. When data Δ = 256 is selected and (1) to (4) are satisfied, the pattern data PD becomes L at all locations in the image, and therefore the conventional bilinear interpolation processing is performed in the entire region of the image. . On the other hand, when the data Δ = 1 is selected and (1) to (4) are established, the pattern data PD is A in all portions of the image, and therefore, processing by the nearest neighbor is performed in the entire region of the image.

  Thus, according to the bilinear filter unit BU, each method of nearest neighbor, bilinear interpolation, and bilinear-nearest neighbor mixed interpolation can be arbitrarily selected by selecting the threshold data Δ. Further, by storing the data Δ in the sprite attribute table 13 as attribute data for each sprite, it becomes possible to use the optimum data Δ for each sprite. Thereby, it is possible to improve the image quality. Further, the bilinear filter unit BU needs only to add a comparator and a decoder as compared with the conventional bilinear filter interpolation circuit, and can improve the conventional circuit at low cost.

  The present invention is used for a display device of a game machine.

It is a block diagram which shows the structure of the bilinear filter unit BU in the image processing apparatus by one Embodiment of this invention. It is a figure for demonstrating the pattern data PD output from the (DELTA) -comparator 31 in FIG. 1 is a block diagram illustrating a configuration of an image processing apparatus according to an embodiment of the present invention. It is a block diagram which shows the structure of the display system using the display system by a sprite. It is a figure for demonstrating a bilinear filter interpolation process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 3 ... Pattern ROM, 4 ... Display apparatus, 11 ... Image processing apparatus, 21 ... Rendering engine, 23 ... Frame buffer, 31 ... (DELTA) -comparator, 32 ... Operation coefficient decoder, 33 ... Bilinear filter interpolation circuit.

Claims (4)

Image processing in the image processing apparatus that displays after the threshold data for determining the color data of the display dot to be displayed with a pre-stored memory means for each sprite pattern, performing a predetermined rendering process on the sprite pattern A method,
First calculation processing for obtaining display coordinates of display dots to be actually displayed as well as obtaining display coordinates of display points for calculation by the rendering processing of the sprite pattern;
A second calculation process of calculating the difference between the color data between each other 4 display point on calculation of the periphery of the display dot the to be actually displayed,
A comparison process for detecting whether the difference is smaller than the threshold value data of the sprite pattern to be displayed stored in the storage means;
Based on the output of the comparison process, among the 4 display point, when the difference of the color data between display points adjacent previous sign of greater than threshold data, among the display points the adjacent be actually displayed Sets and outputs the calculation coefficient that uses the color data of the display point closer to the power display dot as interpolation data, and if it is small, the color data obtained by bilinear interpolation of the color data between adjacent display points is used as the interpolation data A calculation coefficient determination process for setting and outputting a calculation coefficient;
Interpolation processing for obtaining color data of the display dots to be actually displayed using the calculation coefficient output from the calculation coefficient determination process;
An image processing method comprising: Image processing in the image processing apparatus that displays after the threshold data for determining the color data of the display dot to be displayed with a pre-stored memory means for each sprite pattern, performing a predetermined rendering process on the sprite pattern A method,
First calculation processing for obtaining display coordinates of display dots to be actually displayed as well as obtaining display coordinates of display points for calculation by the rendering processing of the sprite pattern;
A second calculation process of calculating the difference between the color data between each other 4 display point on calculation of the periphery of the display dot the to be actually displayed,
There is a pattern determined in advance by the magnitude relationship between each of the color data differences between the four display points and the threshold value data, and the difference obtained by the second calculation process should be stored in the storage means A comparison process for detecting which of the patterns corresponds to whether the sprite pattern is smaller than the threshold value data;
In response to the pattern detected by the comparison process, from the correspondence table of each preset pattern and calculation coefficient, the interpolation method to be applied is selected from nearest neighbor interpolation, bilinear interpolation, and bilinear-nearest neighbor mixed interpolation. A calculation coefficient determination process for obtaining a corresponding calculation coefficient;
Interpolation processing for obtaining color data of the display dots to be actually displayed using the calculation coefficient output from the calculation coefficient determination process;
An image processing method comprising: An image processing apparatus that displays after performing a predetermined rendering process on a sprite pattern,
Storage means for storing threshold data for obtaining color data of display dots to be displayed in advance for each sprite pattern;
First calculation means for obtaining display coordinates of display points to be actually displayed, and calculating display coordinates of display dots to be actually displayed by rendering the sprite pattern;
And second calculating means for calculating a difference between the color data between each other 4 display point on calculation of the periphery of the display dot the to be actually displayed,
Comparing means for detecting whether the difference is smaller than the threshold data of the sprite pattern to be displayed stored in the storage means;
Based on the output of the comparison means, among the 4 display point, when the difference of the color data between display points adjacent previous sign of greater than threshold data, among the display points the adjacent be actually displayed Sets and outputs the calculation coefficient that uses the color data of the display point closer to the power display dot as interpolation data, and if it is small, the color data obtained by bilinear interpolation of the color data between adjacent display points is used as the interpolation data A calculation coefficient decoder that sets and outputs a calculation coefficient;
An interpolation circuit for obtaining color data of the display dots to be actually displayed using the calculation coefficient output from the calculation coefficient decoder;
An image processing apparatus comprising: An image processing apparatus that displays after performing a predetermined rendering process on a sprite pattern,
Storage means for storing threshold data for obtaining color data of display dots to be displayed in advance for each sprite pattern;
First calculation means for obtaining display coordinates of display points to be actually displayed, and calculating display coordinates of display dots to be actually displayed by rendering the sprite pattern;
And second calculating means for calculating a difference between the color data between each other 4 display point on calculation of the periphery of the display dot the to be actually displayed,
There is a predetermined pattern depending on the magnitude relationship between each of the color data differences between the four display points and the threshold data, and the difference obtained by the second computing means is stored in the storage means to be displayed. Comparing means for detecting which of the patterns corresponds to whether the sprite pattern is smaller than the threshold data;
In response to the pattern detected by the comparison means, a preset correspondence table between each pattern and a calculation coefficient is used as an interpolation method to be applied among nearest neighbor interpolation, bilinear interpolation, and bilinear-nearest neighbor mixed interpolation. An arithmetic coefficient decoder for obtaining a corresponding arithmetic coefficient;
An interpolation circuit for obtaining color data of the display dots to be actually displayed using the calculation coefficient output from the calculation coefficient decoder;
An image processing apparatus comprising:

Images